Opstelten Wim, Zuithoff Nicolaas P A, van Essen Gerrit A, van Loon Anton M, van Wijck Albert J M, Kalkman Cornelis J, Verheij Theo J M, Moons Karel G M
Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, P.O. Box 85060, 3500 AB Utrecht, The Netherlands Eijkman Winkler Institute for Microbiology, Inflammation and Infectious Diseases, University Medical Center Utrecht, The Netherlands Pain Clinic, Department of Anaesthesiology, Division Perioperative Care and Emergency Medicine, University Medical Center Utrecht, The Netherlands.
Pain. 2007 Nov;132 Suppl 1:S52-S59. doi: 10.1016/j.pain.2007.02.004. Epub 2007 Mar 26.
Postherpetic neuralgia (PHN) is the most frequent complication of herpes zoster (HZ) and difficult to treat. Timely identification of high-risk HZ-patients enables physicians to focus on PHN prevention. To assess which simple to measure factors are independent predictors of PHN, and whether psychosocial and serological/virological parameters have additional predictive value, a prospective cohort study in primary care was conducted. We included 598 elderly (>50 years) consecutive patients with acute HZ (rash <7 days) below sixth cervical dermatome. At baseline demographic, clinical (e.g., duration and severity of pain and rash), psychological (Pain Cognition List [PCL] and Spielberger's Anxiety Inventory), serological (VZV-antibodies) and virological (viremia presence) variables were measured. Blood tests were performed in a random subset of 218 patients. Primary outcome was significant pain (VAS >30 on 0-100 scale) after three months. The final prediction model obtained from multivariable logistic regression was (internally) validated using bootstrapping techniques, and adjusted for optimism. Forty-six (7.7%) patients developed PHN. Independent predictors were age (odds ratio [OR]=1.08 per year), acute pain severity (OR=1.02 per unit), presence of severe rash (OR=2.31), and rash duration before consultation (OR=0.78 per day): area under receiver-operating-characteristic curve [ROC area]=0.77 (95% CI: 0.71-0.82). Of the five PCL scores, only factor V ('trust in healthcare') was an additional predictor (OR=1.01 per unit), though it increased the ROC area with only 0.01 to 0.78. The Spielberger's anxiety scores and serological and virological variables were no additional predictors. Thus, four simple variables can help physicians to timely identify elderly HZ-patients at risk of PHN.
带状疱疹后神经痛(PHN)是带状疱疹(HZ)最常见的并发症,且难以治疗。及时识别高危HZ患者可使医生专注于预防PHN。为了评估哪些易于测量的因素是PHN的独立预测因素,以及心理社会和血清学/病毒学参数是否具有额外的预测价值,我们在初级保健机构开展了一项前瞻性队列研究。我们纳入了598例年龄大于50岁、急性HZ(皮疹出现时间<7天)且皮疹位于第六颈椎皮节以下的连续患者。在基线时,测量了人口统计学、临床(如疼痛和皮疹的持续时间及严重程度)、心理(疼痛认知清单[PCL]和斯皮尔伯格焦虑量表)、血清学(水痘-带状疱疹病毒抗体)和病毒学(病毒血症情况)变量。对218例患者的随机子集进行了血液检测。主要结局是三个月后出现显著疼痛(视觉模拟评分法[VAS]在0至100分制中>30分)。通过多变量逻辑回归获得的最终预测模型使用自助法进行(内部)验证,并针对乐观偏差进行了调整。46例(7.7%)患者发生了PHN。独立预测因素为年龄(比值比[OR]=每年1.08)、急性疼痛严重程度(OR=每单位1.02)、严重皮疹的存在(OR=2.31)以及就诊前皮疹持续时间(OR=每天0.78):受试者工作特征曲线下面积[ROC面积]=0.77(95%置信区间:0.71 - 0.82)。在五个PCL评分中,只有因素V(“对医疗保健的信任”)是一个额外的预测因素(OR=每单位1.01),尽管它仅将ROC面积增加了0.01至0.78。斯皮尔伯格焦虑评分以及血清学和病毒学变量并非额外的预测因素。因此,四个简单变量可帮助医生及时识别有发生PHN风险的老年HZ患者。